Intent-Guided Reasoning for Sequential Recommendation
Yifan Shao, Peilin Zhou
TL;DR
This work addresses instability and surface-level reasoning in reasoning-enhanced sequential recommendation by anchoring deliberation to high-level user intents. It introduces IGR-SR, comprising a Latent Intent Distiller (LID) to efficiently extract intents, an Intent-aware Deliberative Reasoner (IDR) with a dual-attention flow for intent deliberation and decision-making, and an Intent Consistency Regularization (ICR) with masked augmentation and InfoNCE objectives to ensure robust guidance. The approach relies on a frozen encoder enhanced by learnable prefix tokens and dedicated <intent> tokens, a projection module to align representations, and contrastive learning to maintain cross-view consistency. Empirical results on three public datasets show consistent gains over strong baselines and improved robustness to noisy interactions, with an average improvement of 7.13% and smaller performance drops under 20% behavioral noise.
Abstract
Sequential recommendation systems aim to capture users' evolving preferences from their interaction histories. Recent reasoningenhanced methods have shown promise by introducing deliberate, chain-of-thought-like processes with intermediate reasoning steps. However, these methods rely solely on the next target item as supervision, leading to two critical issues: (1) reasoning instability--the process becomes overly sensitive to recent behaviors and spurious interactions like accidental clicks, and (2) surface-level reasoning--the model memorizes item-to-item transitions rather than understanding intrinsic behavior patterns. To address these challenges, we propose IGR-SR, an Intent-Guided Reasoning framework for Sequential Recommendation that anchors the reasoning process to explicitly extracted high-level intents. Our framework comprises three key components: (1) a Latent Intent Distiller (LID) that efficiently extracts multi-faceted intents using a frozen encoder with learnable tokens, (2) an Intent-aware Deliberative Reasoner (IDR) that decouples reasoning into intent deliberation and decision-making via a dual-attention architecture, and (3) an Intent Consistency Regularization (ICR) that ensures robustness by enforcing consistent representations across different intent views. Extensive experiments on three public datasets demonstrate that IGR-SR achieves an average 7.13% improvement over state-of-the-art baselines. Critically, under 20% behavioral noise, IGR-SR degrades only 10.4% compared to 16.2% and 18.6% for competing methods, validating the effectiveness and robustness of intent-guided reasoning.